I'm a fourth-year PhD student in the Tri-Institutional Program in Computational Biology & Medicine at Cornell, Memorial Sloan Kettering Cancer Center, and The Rockefeller University. I am advised by Quaid Morris and am an NSF Graduate Research Fellow. Previously, I was a research intern at Microsoft Research New England with Alex Lu and Kevin Yang.
I work on representation learning for biological sequences. In particular, my past work has focused on foundation models for mRNA (Orthrus, mRNABench) and I am currently working on foundation models for DNA with MSR. My research leverages knowledge of evolutionary structure to design scalable pretraining objectives that are both sample efficient and highly predictive of downstream biological properties.
NeurIPS, 2026
mRNABench evaluates 75 nucleotide foundation models on 79 mature-mRNA prediction tasks, the only benchmark of its kind. We analyze results through the lens of parameter scaling, pretraining objectives, compressibility of genomic regions, and data splitting strategies.
Nature Methods, 2026
Orthrus leverages splice isoforms and orthology alongside contrastive learning to pretrain a state-of-the-art mRNA foundation model, which is broadly predctive of a wide range of mRNA properties and functions. It is extremely sample and parameter efficient, and can group mRNAs by function in its latent space, the first such demonstration (to our knowledge).
Cell Patterns, 2024
PYPE runs phenome-wide association and Mendelian randomization analyses, annotates variants and genes, and generates standard plots from biobank-scale data.
* Equal contribution.
See Google Scholar or my CV for a complete list.